Plausibility Biases in Learning With Multiple Texts on Controversial Scientific Issues

• In Event: Learning From Text in Today’s Society: Reconciling Multiple Incoherent Sources of Information

Abstract

Objectives
Learners who would like to know more about scientific issues currently debated in public usually read
several texts with conflicting arguments. For example, learners researching information on the Internet on
neuroscientific evidence concerning the existence of a free will encounter a number of documents which
argue for different stances on the issue. In addition, they are likely to possess prior beliefs on the issue
which they can use to evaluate the plausibility of incoming information. The present research was
concerned with the question how these plausibility judgments relate to the comprehension of
controversial science-related texts.
Theoretical framework
Psycholinguistic research on the processing of inconsistencies and educational research on conceptual
change suggest that learners routinely validate the plausibility of incoming information in light of their
current understanding and prior knowledge (epistemic validation). We propose that by default,
information which is consistent with learner’s beliefs is more likely to be integrated into the mental
representation (situation model) of controversial scientific issue, whereas inconsistent information tends
to be rejected (plausibility bias, Schroeder, Richter, & Hoever, 2008). However, learners motivated to
develop an adequate and justified point of view can engage in strategic and effortful epistemic elaboration
of information which they find implausible at first sight. As a result, they will be able to develop a more
balanced mental representation of the controversial issue.
Method
We tested these hypotheses in two experiments with sixty-three university students (Experiment 1) and
seventy-five students in upper secondary school (age 16-18 years, Experiment 2). Participants read four
texts that represented two contrary positions on one of two controversial scientific issues (potential of
biomass as energy source/existence of a free will). They were instructed to follow an epistemic reading
goal (develop a justified point of view!) or a receptive reading goal (memorize information!). After each
text, participants completed a recognition task with inference items (situation model), paraphrase items
(memory for text), and distracters items, and provided plausibility judgments for the same set of items.
Results
We used multilevel models (test items nested within persons) with a logit-link function to investigate the
relationship of plausibility judgments and comprehension on a trial-by-trial basis. In both experiments,
results revealed a large plausibility bias in the construction of a situation model of the scientific
controversy. In addition, perceived plausibility was associated with better memory for text and falsepositive
classifications of unrelated information as coming from the text (illusory comprehension).
However, an epistemic learning goal attenuated the plausibility bias in situation model construction
compared to a receptive learning goal.
Scientific significance
Our results support the idea that epistemic validation plays a major role in learning with multiple texts on
controversial scientific issues. What learners will understand and remember of a scientific controversy is
strongly related to what they find plausible. Most important, a plausibility bias can lead to an
impoverished and one-sided representation of controversial issues, but learners can actively work against
this plausibility bias by pursuing an epistemic learning goal.

Authors

• Tobias Richter, University of Kassel

• Johanna Maier, University of Kassel